Sage Journals: Discover world-class research

Abstract

Background: Medication non-adherence is common in Parkinson’s disease (PD) and is associated with increased disability and healthcare costs. Individuals’ beliefs regarding their medical conditions and treatments impact medication adherence. While instruments exist to measure patients’ beliefs about medications in general, no such tool exists for PD.

Objectives: Create an instrument eliciting medication beliefs of persons with PD; identify demographic and clinical characteristics associated with beliefs; and examine whether beliefs are associated with dopaminergic therapy adherence.

Methods: We developed the Parkinson’s Disease Medication Beliefs Scale (PD-Rx) in four phases: focus groups of patients and caregivers to generate items, scale development, expert and patient revision of items, and a cross-sectional validation sample (n = 75). Adherence was calculated using two approaches incorporating self-reported medication lists.

Results: The PD-Rx consists of 11 items covering benefits and risks of PD pharmacotherapies. The scale covers motor improvement, current adverse effects, and future concerns. Higher scores indicate more positive beliefs. Internal consistency was acceptable (Cronbach’s alpha = 0.67). Test-retest reliability was 0.47. Quality of life was associated with PD-Rx scores, and lower scores were associated with non-adherence.

Conclusions: Negative beliefs about PD treatments are associated with lower quality of life and may be related to medication non-adherence. Further study of any causal relationship between beliefs and medication non-adherence in PD will inform the design of future patient-centered interventions to improve adherence.

Keywords

Patient beliefs psychometric adherence health belief model Parkinson’s disease

INTRODUCTION

Despite the efficacy of pharmacotherapy in Parkinson’s disease (PD), adherence to antiparkinsonian drugs is as low as 33% by electronic medication monitoring technology, which is much lower than self-report [1, 2]. Suboptimal adherence with PD medications is associated with worsening disability, higher hospitalization rates, and greater healthcare expenditures [2, 3].

The health belief model (HBM) provides a framework for exploring characteristics influencing health behaviors such as adherence. The HBM proposes that health behavior results from the interaction of three factors: 1) the existence of a salient health concern; 2) the perceived threat of that concern to one’s wellbeing; and 3) the belief that a particular behavior would reduce the threat at an acceptable cost. Here, cost encompasses both financial burden and the effort required to overcome perceived barriers [4].

Critical to an individual’s perceived cost-benefit analysis of treatment are the beliefs one holds about the therapy itself. Medication beliefs may be grouped into two categories: 1) benefits of a treatment versus 2) the risks of said treatment or concerns held by the individual [5]. Instruments evaluating medication beliefs have been validated in multiple conditions, however, prior studies of patient-held beliefs in PD have described unique medication and symptom-specific phobias that are absent from such instruments. For example, patients often endorse fears surrounding the use of levodopa and the risk of developing dyskinesias [6 –9]. Such fears may significantly influence adherence [6 , 13].

Adherence is also associated with demographic and clinical characteristics such as age, education, health literacy, neuropsychiatric symptoms, disease severity, quality of life, polypharmacy, and medication cost in non-PD cohorts [11 –20]. Emerging evidence suggests that these characteristics shape patients’ beliefs about their health and treatments, and such beliefs form a potentially actionable obstacle to adherence [5 , 21–25]. Figure 1 illustrates our conceptual model to explain the characteristics shaping adherence in PD, with beliefs as an intermediary. Our objectives were to develop an instrument and conduct an initial pilot study to capture patient beliefs about PD medications, explore the associations of demographic and clinical characteristics with these beliefs, and define the relationship between beliefs and adherence. We hypothesized that older age, low health literacy, depression, apathy, and poor quality of life would be associated with negative medication beliefs in PD. We further postulated that positive beliefs would predict higher adherence.

MATERIALS AND METHODS

Item generation, scale development, revision, and psychometric testing were accomplished in the following phases: initial focus groups for item generation and scale development, expert review of items, a second round of focus groups for content validity, and psychometric testing in a separate validation cohort. The study was approved by the University of Pennsylvania Institutional Review Board.

Subjects

Focus groups

We conducted 6 focus groups that consisted of a purposive sample of patients with PD or caregivers of patients drawn from existing local PD supportgroups.

Validation cohort

Subjects in the validation cohort were drawn from an existing longitudinal cohort study, the University of Pennsylvania Udall Center for Parkinson’s Research. These subjects had a diagnosis of PD based on established criteria, were 50 years of age or older, and had a score of ≥20 on the Montreal Cognitive Assessment (MoCA) at the time of the validation study visit [26, 27]. Subjects in the validation cohort did not participate in the focus groups.

Item generation and scale development

We employed the nominal group technique (NGT) for item generation [28]. After a warm-up exercise, participants went through cycles of NGT for 3 separate stimulus questions. The questions were: how PD had affected participants’ lives, the key benefits of the dopaminergic medications taken for PD— either perceived or experienced; and the key risks of dopaminergic medications. Three initial focus groups were held, after which individual responses to the benefit and risk prompts were categorized into thematic groupings. The most common themes were converted to 24 items and scored on a 5-point Likert scale, ranging from “strongly disagree” (1 point) to “strongly agree” (5 points).

Item revision

The draft instrument and a standardized review tool were sent to 19 movement disorders experts across different institutions. Experts commented on readability, comprehensiveness, applicability, and missing themes. The modified draft was then presented to a second round of focus groups for revision based on similar criteria. Based on these responses, the instrument wording was revised to include lay explanations of symptoms, and named the PD Medication Beliefs (PD-Rx) instrument.

Psychometric testing

The PD-Rx was administered to a cross-sectional sample of non-demented subjects with PD drawn from the Udall Center. We enrolled subjects who presented for an initial enrollment or annual follow-up visit between May 2013 and May 2014.

Primary variables of interest

In addition to data collected as part of the Udall study (described in detail elsewhere [29, 30] and briefly below), the PD-Rx was administered along with the Brief Medication Questionnaire (BMQ). The BMQ is a validated instrument to measure adherence [31]. The BMQ asks for each medication taken in the past week, including the name, strength, frequency of daily doses, number of doses missed, and indication. Subsequent questions ask whether the subject experiences specific problems with his medications (side effects, trouble remembering all doses, trouble affording medications, etc.). To assess test-retest reliability of the PD-Rx, a second copy of the instrument was mailed to all subjects within two weeks of their study visit.

We assessed adherence in two ways. First, the BMQ Adherence Risk Score (ARS) as mentioned above [31]. The ARS is scored categorically 0–4, with points given for 1) non-adherence with current drug regimen, 2) negative beliefs or motivational barriers, 3) recall barriers, and 4) access barriers. Second, we reviewed the subject’s electronic medical record to obtain the medication list documented by the subject’s neurologist from the most recent clinical visit or telephone encounter immediately preceding the study visit. All dopaminergic medications were converted to levodopa-equivalent doses (LEDs) [32]. Chart review was conducted while blinded to BMQ and PD-Rx responses. Each physician-listed medication was compared to the subject-listed medications in the BMQ. The difference between the subject and physician total LED yielded a percentage LED adherence, with <80% or >120% defined as non-adherent.

Covariates of interest

Demographic and clinical characteristics included age at PD diagnosis and duration of symptoms. The Unified Parkinson’s Disease Rating Scale (UPDRS) was used to capture disease severity. The Hoehn and Yahr scale was used to classify the stage of PD [35]. Apathy was scored from 0–4 based on one UPDRS item, where 4 indicates complete loss of initiative [34]. The 15-item Geriatric Depression Scale was used to measure depressive symptoms, with scores ≥6 indicating depressed mood [36]. Quality of life was assessed using the Parkinson’s Disease Questionnaire (PDQ-39) scored from 0–100 where higher scores denote poorer quality of life [37]. Health literacy was assessed with the Newest Vital Sign (NVS), a brief screening instrument that requires no handwriting and is untimed [38].

Statistical analysis

Item response frequencies were reviewed for non-response, “not applicable” answers, and floor and ceiling effects for item reduction [39]. Cronbach’s alpha was calculated to estimate internal consistency of the scale. Exploratory factor analysis was used to examine construct validity and identify unique domains [40]. Test-retest reliability was assessed using an intra-class correlation coefficient. Univariate analyses were conducted between the PD-Rx total score and individual covariates, using chi-square, t-tests, or ANOVA, as appropriate. Multivariable linear regression analyses were conducted using backwards stepwise elimination with PD-Rx total score as the dependent variable, and any covariates with p-value <0.2 in univariate analyses. Model fit was assessed by likelihood ratio testing.

The association between PD-Rx and ARS was modeled using ordered logistic regression. The proportional odds assumption was tested via likelihood ratio testing. The association between PD-Rx score and binary LED adherence was examined using logistic regression. All statistical analyses were two-sided with significance set at <0.05 level. Data analysis was performed using STATA 12.1 (StataCorp, College Station, TX).

RESULTS

Sample characteristics

Six focus groups ranged in size from 7 to 16 members for a total of 38 participants; 71% were patients and 29% caregivers. A total of 75 subjects participated in the validation cohort; their characteristics are displayed in Table 1. Significant apathy or depressed mood at the time of the study visit was uncommon, however 28% of the sample endorsed antidepressant use. By ARS criteria, 81.3% of subjects had at least two out of four barriers to adherence.

Item generation and scale development

Initial items fell into several themes: 1) positive themes, including improvements in motor and non-motor symptoms, and restoration of normalcy; 2) negative themes such as known adverse effects, disruption of daily activity due to regimen complexity, and financial burden; and 3) future concern themes of tachyphylaxis, “running out of options”, and neuroprotection.

Expert reviewers recommended the removal of 7 original items, yielding 17 items in the pilot PD-Rx, none of which had >5% missing data. Item-level distribution of scores ranged from 1–5 for all except for one item. After factor analysis 3 items not specific to PD were deleted, as were 3 items with≥20% “not applicable” responses. Table 2 shows the final 11-item PD-Rx scale and the percent choosing each option. The total possible score ranges from 11–55, with higher scores indicating benefits outweighing risks, or more positive medication beliefs.

Reliability and validity

PD-Rx scores were normally distributed, with a mean of 38.9 (SD 4.9) and range of 29–51. Cronbach’s alpha was 0.67.The final 11 items loaded onto 3 factors— motor improvement/self-efficacy, current adverse effects, and future concerns. As the 3 factors explained only 58.3% of score variance, factor-based scoring was rejected in favor of an overall score. Test-retest reliability based on a subset of 56 subjects was 0.47 for the total score, and ranged from 0.00 to 0.60 for the individual items, as shown in Table 3.

Associations of PD-Rx scores with covariates of interest

Table 4 shows unadjusted associations of medication beliefs with the covariates of interest. Depressed mood and poorer quality of life were inverselycorrelated with PD-Rx scores (p = 0.02 and p < 0.01, respectively). There was no association between PD-Rx score and the number of PD medications or total number of medications taken (p = 0.76, and 0.85, respectively).

Table 5 displays the associations of individual covariates after adjusting for demographics and disease stage. Depressed mood was not independently associated with medication beliefs (p = 0.56), however the relationship between poor quality of life and negative beliefs remained significant (–0.13 points on PD-Rx for every point on PDQ-39, p = 0.04).

Associations of PD-Rx Scores with Adherence Measures

Using the ARS, with higher scores indicative of non-adherence, every point on the PD-Rx was associated with a 0.87-fold lower odds of non-adherence (95% CI 0.79–0.95, p < 0.01). This indicates that benefit-predominant beliefs are associated with fewer barriers to adherence. The association between PD-Rx and binary adherence, defined as 80–120% LED concordance, trended to significance, with an odds ratio of 1.11 for every point on the PD-Rx (95% CI 0.99–1.25, p = 0.07).

DISCUSSION

We identified medication beliefs held by diverse populations of patients with PD and their caregivers and coalesced 11 common beliefs about dopaminergic medications covering both necessities and concerns into a patient-centered instrument with reasonably good psychometric characteristics. Poorer quality of life was associated with a lower total score on the PD-Rx (more negative beliefs), and lower scores were also related to at least one measure of adherence.

Two surprising findings in our study were the low overall test-retest reliability of the PD-Rx total score (0.47) and the highly variable reliability of individual items. Certain items were particularly susceptible to change; in particular, prompts such as “Taking my PD medications improves how fast or easily I move” or “Taking my PD medications makes walking easier”— had very low intra-class correlation coefficients. We hypothesize that subjects responded based on their functional status at the moment of instrument completion. As bradykinesia, rigidity, and certain components of gait are sensitive to on- and off-medication states that may fluctuate throughout the course of a day, responses to such questions may indicate a direct medication effect. Other items appear to probe more stable, non-state-dependent beliefs, such as offering some control over PD or slowing its progression. Therefore, an individual’s subjective weighting of necessities versus concerns may vary over the course of a single day in parallel with PD motor fluctuations. The alternative explanation is that the items are not adequately capturing the domains we believe them to be. Further revisions of the PD-Rx will include an item indicating “on” or “off” medication state, and we will examine the test-retest reliability between fluctuators and non-fluctuators to determine whether medication states affect the reporting of beliefs.

Both depressed mood and poorer quality of life were associated with lower PD-Rx scores, and poorer quality of life remained independently associated with negative beliefs after adjusting for confounders. This supports the hypothesis that medication beliefs may be a mediator between clinical characteristics and adherence. By identifying a patient with depressed mood or poor quality of life as potentially non-adherent, the clinician takes the first step towards intervening on behavior. However, without understanding the specific beliefs motivating that individual’s behavior, one-size-fits-all interventions will remain the suboptimal standard.

Other covariates associated with non-adherence in prior work— such as education, racial minority, age, and low health literacy— were not significantly correlated with beliefs in the current study. For the former three, we may have failed to detect an association due to having a highly educated, nearly entirely white sample without much variability in age. Health literacy was also not related to beliefs. Prior studies finding that successful adherence required adequate health literacy relied upon pharmacy refill records as their measure of adherence, and our choice of adherence measures may have biased our results towards the null [17].

Although we hypothesized that apathy and depressed mood might bias the estimation of risk in PD treatment, neither characteristic was independently associated with beliefs. This may be due to the lower than expected prevalence of both apathy and depressed mood in our sample [41]. However, 28% of our sample was treated with an antidepressant, suggesting that pharmacologic treatment blunted our ability to detect a signal [42].

Because of the exploratory nature of this study, we relied on self-report and physician-documented medication lists to calculate PD medication adherence. Similar to prior PD studies relying on self-report, we found 25% of our sample was non-adherent by LED criteria [2, 43]. This is likely an underestimate. Studies using pharmacy refill data and pill counts have detected non-adherence approaching 67% and higher [43]. Therefore our results are likely biased towards the null.

This exploratory study has several limitations. First, there are potential sampling and healthy volunteer biases. One attempt to overcome this limitation was to implement in-home study visits for subjects with limited mobility. However, this initial pilot was conducted with a small and demographically homogeneous sample and may be underpowered despite recruiting at least five subjects per item. Second, focus groups bear the risk of social desirability bias. We addressed this by drawing our focus group participants from pre-existing support groups [44]. As mentioned above, patient-reported medication lists may significantly underestimate the prevalence of non-adherence, and provider-reported medication lists drawn from the electronic medical record are prone to errors [45, 46]. Future studies employing gold-standard adherence measures, such as electronic medication monitoring devices, will be critical for validation of the PD-Rx. Finally, whether further validation proves an association between beliefs and adherence, the possibility that such beliefs may be refractory to intervention must be considered.

In summary, we present the initial pilot of an instrument to assess the medication beliefs held by patients with PD. With the PD-Rx, we sought to identify the beliefs that underlie medication phobia and non-adherence. Similar to providers directly asking patients about adherence, we felt that directly asking patients about fear of medication would lead to biased responses [47]. Further research is indicated to determine the effect of on- and off-medication states on beliefs, to test the PD-Rx in different samples, to confirm the association between beliefs and adherence using more robust measures, and to develop targeted interventions. Non-adherent PD patients face significantly higher rates of acute care visits and annual medical costs, even after adjusting for comorbidities [43]. Understanding the patient and disease-specific factors influencing adherence could guide clinicians to individualize counseling and patient education, though this remains to be tested in PD [48].

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

AUTHORSHIP

JE Fleisher: Conception and design of study, acquisition of data, analysis and interpretation of data; drafting and revision of manuscript; final approval of manuscript.

NA Dahodwala: Conception and design of study, analysis and interpretation of data; drafting and revision of manuscript; final approval of manuscript.

M Mayo: Acquisition and analysis of data; revision of manuscript; final approval of manuscript.

SX Xie: Conception and design of study, analysis and interpretation of data; drafting and revision of manuscript; final approval of manuscript.

D Weintraub: Interpretation of data; revision of manuscript; final approval of manuscript.

J Chodosh: Interpretation of data; revision of manuscript; final approval of manuscript.

JA Shea: Conception and design of study, analysis and interpretation of data; drafting and revision of manuscript; final approval of manuscript.

Footnotes

ACKNOWLEDGMENTS

Data contributed to this project by the MorrisK. Udall Center at the Perelman School of Medicine at the University of Pennsylvania (P50 NS053488, Trojanowski JQ-PI; Weintraub D-Core B Leader; Chen-Plotkin A-Investigator and Project 1 PI, Chahine L-Investigator; Duda JE-Investigator; Dahodwala, N-Investigator; Rick J-Project-Project Manager). The authors thank Bonnie Svarstad, PhD for permission to use the Brief Medication Questionnaire in this study, and the patients and caregivers for their participation. Dr. Fleisher received support from NIH T32-NS-061779. Dr. Dahodwala is supported by NIA K23 AG034236. The Udall Center of Excellence is funded by a grant from NINDS (NS-053488).

Appendix

References

Grosset

, Bone

, Reid

, & Grosset

(2006) Measuring therapy adherence in Parkinson’s disease: A comparison of methods, J Neurol Neurosurg Psychiatry 77, 249–251.

Kulkarni

, Balkrishnan

, Anderson

, Edin

, Kirsch

, & Stacy

(2008) Medication adherence and associated outcomes in medicare health maintenance organization-enrolled older adults with Parkinson’s disease, Mov Disord 23, 359–365.

Grosset

, Antonini

, Canesi

, Pezzoli

, Lees

, Shaw

, Cubo

, Martinez-Martin

, Rascol

, Negre-Pages

, Senard

, Schwarz

, Strecker

, Reichmann

, Storch

, Lohle

, Stocchi

, & Grosset

(2009) Adherence to antiparkinson medication in a multicenter European study, Mov Disord 24, 826–832.

Rosenstock

, Strecher

, & Becker

(1988) Social learning theory and the Health Belief Model, Health Educ Q 15, 175–183.

Horne

, & Weinman

(1999) Patients’ beliefs about prescribed medicines and their role in adherence to treatment in chronic physical illness, J Psychosom Res 47, 555–567.

Schlesinger

, & Rabinowitz

(2011) Medication phobia: A new cause of drug noncompliance in Parkinson disease, Clin Neuropharmacol 34, 220–223.

Vlaar

, Hovestadt

, van Laar

, & Bloem

(2011) The treatment of early Parkinson’s disease: Levodopa rehabilitated, Pract Neurol 11, 145–152.

Fujimoto

, Murata

, Hattori

, & Kondo

(2011) [Patients’ perspective on Parkinson disease therapies: Results of a large-scale survey in Japan], Brain Nerve 63, 255–265.

Hung

, Adeli

, Arenovich

, Fox

, & Lang

(2010) Patient perception of dyskinesia in Parkinson’s disease, J Neurol Neurosurg Psychiatry 81, 1112–1115.

10.

Horne

, Chapman

, Parham

, Freemantle

, Forbes

, & Cooper

(2013) Understanding patients’ adherence-related beliefs about medicines prescribed for long-term conditions: A meta-analytic review of the Necessity-Concerns Framework, PLoS One 8, e80633.

11.

Daley

, Myint

, Gray

, & Deane

(2012) Systematic review on factors associated with medication non-adherence in Parkinson’s disease, Parkinsonism Relat Disord 18, 1053–1061.

12.

Bainbridge

, & Ruscin

(2009) Challenges of treatment adherence in older patients with Parkinson’s disease, Drugs Aging 26, 145–155.

13.

DiMatteo

, Lepper

, & Croghan

(2000) Depression is a risk factor for noncompliance with medical treatment: Meta-analysis of the effects of anxiety and depression on patient adherence, Arch Intern Med 160, 2101–2107.

14.

Insel

, Morrow

, Brewer

, & Figueredo

(2006) Executive function, working memory, and medication adherence among older adults, J Gerontol B Psychol Sci Soc Sci 61, P102–P107.

15.

Maguire

, Hughes

, & McElnay

(2008) Exploring the impact of depressive symptoms and medication beliefs on medication adherence in hypertension–a primary care study, Patient Educ Couns 73, 371–376.

16.

Gellad

, Grenard

, & Marcum

(2011) A systematic review of barriers to medication adherence in the elderly: Looking beyond cost and regimen complexity, Am J Geriatr Pharmacother 9, 11–23.

17.

Gazmararian

, Kripalani

, Miller

, Echt

, Ren

, & Rask

(2006) Factors associated with medication refill adherence in cardiovascular-related diseases: A focus on health literacy, J Gen Intern Med 21, 1215–1221.

18.

Vermeire

, Hearnshaw

, Van Royen

, & Denekens

(2001) Patient adherence to treatment: Three decades of research. A comprehensive review, J Clin Pharm Ther 26, 331–342.

19.

Panos

, Del Re

, Thames

, Arentsen

, Patel

, Castellon

, Singer

, & Hinkin

(2014) The impact of neurobehavioral features on medication adherence in HIV: Evidence from longitudinal models, AIDS Care 26, 79–86.

20.

Kale

, Federman

, Krauskopf

, Wolf

, O’Conor

, Martynenko

, Leventhal

, & Wisnivesky

(2015) The association of health literacy with illness and medication beliefs among patients with chronic obstructive pulmonary disease, PLoS One 10, e0123937.

21.

Graham

, Bennett

, Holmes

, & Gross

(2007) Medication beliefs as mediators of the health literacy-antiretroviral adherence relationship in HIV-infected individuals, AIDS Behav 11, 385–392.

22.

Figueiras

, Alves

, Marcelino

, Cortes

, Weinman

, & Horne

(2009) Assessing lay beliefs about generic medicines: Development of the generic medicines scale, Psychol Health Med 14, 311–321.

23.

French

, James

, Horne

, & Weinman

(2005) Causal beliefs and behaviour change post-myocardial infarction: How are they related?, Br J Health Psychol 10, 167–182.

24.

Horne

, Buick

, Fisher

, Leake

, Cooper

, & Weinman

(2004) Doubts about necessity and concerns about adverse effects: Identifying the types of beliefs that are associated with non-adherence to HAART, Int J STD AIDS 15, 38–44.

25.

Horne

, Graupner

, Frost

, Weinman

, Wright

, & Hankins

(2004) Medicine in a multi-cultural society: The effect of cultural background on beliefs about medications, Soc Sci Med 59, 1307–1313.

26.

Hughes

, Daniel

, Kilford

, & Lees

(1992) Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases, J Neurol Neurosurg Psychiatry 55, 181–184.

27.

Chou

, Amick

, Brandt

, Camicioli

, Frei

, Gitelman

, Goldman

, Growdon

, Hurtig

, Levin

, Litvan

, Marsh

, Simuni

, Troster

, Uc

, & Parkinson Study Group Cognitive/Psychiatric Working Group (2010) A recommended scale for cognitive screening in clinical trials of Parkinson’s disease, Mov Disord 25, 2501–2507.

28.

Gallagher

, Hares

, Spencer

, Bradshaw

, & Webb

(1993) The nominal group technique: A research tool for general practice?, Fam Pract 10, 76–81.

29.

Morley

, Xie

, Hurtig

, Stern

, Colcher

, Horn

, Dahodwala

, Duda

, Weintraub

, Chen-Plotkin

, Van Deerlin

, Falcone

, & Siderowf

(2012) Genetic influences on cognitive decline in Parkinson’s disease, Mov Disord 27, 512–518.

30.

Watson

, Cholerton

, Gross

, Weintraub

, Zabetian

, Trojanowski

, Montine

, Siderowf

, & Leverenz

(2013) Neuropsychologic assessment in collaborative Parkinson’s disease research: A proposal from the National Institute of Neurological Disorders and Stroke Morris K. Udall Centers of Excellence for Parkinson’s Disease Research at the University of Pennsylvania and the University of Washington, Alzheimers Dement 9, 609–614.

31.

Svarstad

, Chewning

, Sleath

, & Claesson

(1999) The Brief Medication Questionnaire: A tool for screening patient adherence and barriers to adherence, Patient Educ Couns 37, 113–124.

32.

Tomlinson

, Stowe

, Patel

, Rick

, Gray

, & Clarke

(2010) Systematic review of levodopa dose equivalency reporting in Parkinson’s disease, Mov Disord 25, 2649–2653.

33.

Goetz

, Fahn

, Martinez-Martin

, Poewe

, Sampaio

, Stebbins

, Stern

, Tilley

, Dodel

, Dubois

, Holloway

, Jankovic

, Kulisevsky

, Lang

, Lees

, Leurgans

, LeWitt

, Nyenhuis

, Olanow

, Rascol

, Schrag

, Teresi

, Van Hilten

, & LaPelle

(2007) Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Process, format, and clinimetric testing plan, Mov Disord 22, 41–47.

34.

Goetz

, Tilley

, Shaftman

, Stebbins

, Fahn

, Martinez-Martin

, Poewe

, Sampaio

, Stern

, Dodel

, Dubois

, Holloway

, Jankovic

, Kulisevsky

, Lang

, Lees

, Leurgans

, LeWitt

, Nyenhuis

, Olanow

, Rascol

, Schrag

, Teresi

, van Hilten

, LaPelle

, & Movement Disorder Society URTF (2008) Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Scale presentation and clinimetric testing results, Mov Disord 23, 2129–2170.

35.

Goetz

, Poewe

, Rascol

, Sampaio

, Stebbins

, Counsell

, Giladi

, Holloway

, Moore

, Wenning

, Yahr

, Seidl

, & Movement Disorder Society Task Force on Rating Scales for Parkinson’s Disease (2004) Movement Disorder Society Task Force report on the Hoehn and Yahr staging scale: Status and recommendations, Mov Disord 19, 1020–1028.

36.

Sheikh

, & Yesavage

(1986) Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version, Clin Gerontol 5, 165–172.

37.

Jenkinson

, Fitzpatrick

, Peto

, Greenhall

, & Hyman

(1997) The Parkinson’s Disease Questionnaire (PDQ-39): Development and validation of a Parkinson’s disease summary index score, Age Ageing 26, 353–357.

38.

Osborn

, Weiss

, Davis

, Skripkauskas

, Rodrigue

, Bass

, & Wolf

(2007) Measuring adult literacy in health care: Performance of the newest vital sign, Am J Health Behav 31(Suppl 1), S36–S46.

39.

Streiner

, & Norman

(2008), Health Measurement Scales: A Practical Guide to Their Development and Use, Oxford University Press, Oxford.

40.

Streiner

(1994) Figuring out factors: The use and misuse of factor analysis, Can J Psychiatry 39, 135–140.

41.

Pedersen

, Larsen

, Alves

, & Aarsland

(2009) Prevalence and clinical correlates of apathy in Parkinson’s disease: A community-based study, Parkinsonism Relat Disord 15, 295–299.

42.

Aarsland

, Marsh

, & Schrag

(2009) Neuropsychiatric symptoms in Parkinson’s disease, Mov Disord 24, 2175–2186.

43.

Davis

, Edin

, & Allen

(2010) Prevalence and cost of medication nonadherence in Parkinson’s disease: Evidence from administrative claims data, Mov Disord 25, 474–480.

44.

Dahodwala

, Siderowf

, Xie

, Noll

, Stern

, & Mandell

(2009) Racial differences in the diagnosis of Parkinson’s disease, Mov Disord 24, 1200–1205.

45.

Manley

, Drayer

, McClaran

, Bender

, & Muther

(2003) Drug record discrepancies in an outpatient electronic medical record: Frequency, type, and potential impact on patient care at a hemodialysis center, Pharmacotherapy 23, 231–239.

46.

Wagner

, & Hogan

(1996) The accuracy of medication data in an outpatient electronic medical record, J Am Med Inform Assoc 3, 234–244.

47.

Valldeoriola

, Coronell

, Pont

, Buongiorno

, Camara

, Gaig

, Compta

, & Group

(2011) Socio-demographic and clinical factors influencing the adherence to treatment in Parkinson’s disease: The ADHESON study, Eur J Neurol 18, 980–987.

48.

Clifford

, Barber

, & Horne

(2008) Understanding different beliefs held by adherers, unintentional nonadherers, and intentional nonadherers: Application of the Necessity-Concerns Framework, J Psychosom Res 64, 41–46.

Development and Validation of the Parkinson’s Disease Medication Beliefs Scale (PD-Rx)

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Subjects

Focus groups

Validation cohort

Item generation and scale development

Item revision

Psychometric testing

Primary variables of interest

Covariates of interest

Statistical analysis

RESULTS

Sample characteristics

Item generation and scale development

Reliability and validity

Associations of PD-Rx scores with covariates of interest

Associations of PD-Rx Scores with Adherence Measures

DISCUSSION

CONFLICT OF INTEREST

AUTHORSHIP

Footnotes

ACKNOWLEDGMENTS

Appendix

References